ESRL Quarterly Newsletter - Spring 2010

GPS for Storms, Climate, More

ESRL sends innovative instrument packages offshore for first time

In preparation for this year's hurricane season, several ESRL researchers spent the winter working with Devon Energy, placing Frisbee-sized satellite antennas high atop two Devon oil platforms in the Gulf of Mexico. The instruments are based on Global Positioning System technology- most commonly known for its precise location measurements.

It's not as if storms are likely to push around the oil platforms. Rather, ESRL researchers are testing their ability to use a serendipitous power of GPS signals- information that can be gleaned about atmospheric water vapor- to improve regional weather forecasts, monitor climate, and help verify satellite and balloon measurements. The Gulf installations of "GPS-Meteorology" packages are the first deployed over open water far enough from land that satellite observations are not "contaminated" by land interference.

That could mean better hurricane intensity forecasts- since moisture-rich air can drive the intensification of a storm, and dry air can weaken it. But the GPS-Met packages may improve weather forecasting much farther away.

"If successful, the experiment could have far-reaching consequences, since most of the atmospheric moisture in the eastern two-thirds of the United States has its origins in the Gulf of Mexico," said ESRL's Seth Gutman (Global Systems Division).

Onshore weather forecasts will likely see immediate improvement, Gutman said, but the GPS-Met data may also help researchers assess satellite measurements of phenomena important to weather and climate- from the amount of water in the air to the measurement of sea level.

Gutman leads a team of ESRL researchers who have spent the last decade figuring out how to extract measurements of atmospheric water vapor from normal GPS signals, and how to assimilate those data into weather and other Earth system models.

By any measure, the team's success has been remarkable.

Gutman's team has documented how GPS water vapor measurements led to better severe weather forecasts in California, Canada, the U.S. Southeast, and other regions.

"In 10 years, our ability to measure the amount of water vapor in the atmosphere has improved by about 53 percent," Gutman said.

Water vapor is notoriously difficult to measure and understand, and observations using satellites, weather balloons, lidars, and microwave instruments are also often costly. Scientists seeking to understand weather and climate, however, need reliable measurements- atmospheric water vapor is the means by which moisture and latent heat are moved around the planet, causing weather. Water is also a greenhouse gas, affecting heat balance both directly and through its role in cloud formation.

In the early 1990s, scientists realized they could mathematically parse out total atmospheric water vapor from dual-frequency GPS satellite radio signals. In essence, water vapor bends and slows GPS radio signals traveling from satellites to ground-based GPS receivers. Researchers can measure the delay from several GPS satellites simultaneously and use math to extract the role- and amount- of water.

There are other sources of data on atmospheric water content, Gutman said, perhaps most importantly, NOAA satellite data. Those measurements, however, contain significant uncertainties, because some of the satellite instruments can "see" only over the oceans, other instruments can't see through clouds.

GPS satellites broadcast nearly continuously and in theory, water vapor information can be extracted from the hundreds of GPS ground receivers already serving other functions - earthquake detection and surveying, for example. Currently, NOAA is regularly importing data from about 300 GPS sites, Gutman said, and to transition GPS-Met into operations would require about 800 sites. The data are pulled into a system Gutman and his colleagues built to assimilate observations into operational weather models. When the system went live in 2005, the results were instant.

"You could see the system error drop instantly - Boom!- when we used these data," Gutman said. "The models were just starved for water vapor information."

Today, he's working closely with the National Weather Service Lake Charles, LA Forecast Office, to assimilate data from the two new Gulf of Mexico platforms, and his team plans to document the subsequent change (likely significant improvement) in forecast skill.

They are also collaborating with Louisiana State University researchers, who want to use GPS-Met data to make more accurate measurements of water height changes associated with storm surges, subsidence, and climate change.

In collaborative work with NOAA's National Environmental Satellite, Data, and Information Service and the Cooperative Institute for Mesoscale Meteorological Studies, the team used GPS-Met data to identify an unexpected discrepancy between the GOES East and West satellite measurements, and figured out how to correct for it.

Finally, the ESRL team is working to install GPS-Met packages in other critical locations, including NOAA's climate observatories. When a station was installed in Barrow, AK recently, Gutman's team found disagreement between his instrument's measurements and observations made by balloon-borne instruments. It turned out that certain sensors on the balloon package were faulty.

"These problems were going unnoticed, because there was nothing to compare the data with," Gutman said. "Now we have a reliable point of comparison."